Efficient Gene Expression System Using the RTP801 Promoter in the Corpus Cavernosum of High-Cholesterol Diet-Induced Erectile Dysfunction Rats for Gene Therapy

Efficient Gene Expression System Using the RTP801 Promoter inthe Corpus Cavernosum of High-Cholesterol Diet-Induced ErectileDysfunction Rats for Gene Therapy

Minhyung Lee, PhD,* Ji-Kan Ryu, MD, PhD,† Shuguang Piao, MD,† Min Ji Choi, MS,†

Hyun Ah Kim, BS,* Lu-Wei Zhang, MD,† Hwa-Yean Shin, MS,† Haeng In Jung, MS,‡

In-Hoo Kim, MD, PhD,‡ Sung Wan Kim, PhD,§ and Jun-Kyu Suh, MD, PhD†

*Department of Bioengineering, College of Engineering, Hanyang University, Seoul, Korea; †Department of Urology, InhaUniversity School of Medicine, Incheon, Korea; ‡Research Institute, National Cancer Center, Goyang, Gyeonggi, Korea;§Center for Controlled Chemical Delivery, Department of Pharmaceutics and Pharmaceutical Chemistry, University ofUtah, Salt Lake City, UT, USA

DOI: 10.1111/j.1743-6109.2008.00771.x

A B S T R A C T

Introduction. The application of gene therapy for a nonlife-threatening disease, such as erectile dysfunction (ED),requires a higher safety level and more efficacious systems for gene transfer.Aim. To establish a novel technique for gene expression in a rat model of hypercholesterolemic ED that uses theRTP801 promoter, a hypoxia-inducible promoter.Methods. Two-month-old male Sprague–Dawley rats were fed a diet containing 4% cholesterol and 1% cholic acid,and age-matched control animals were fed a normal diet, for 3 months.Main Outcome Measures. Cavernous expression of hypoxia-inducible factor (HIF)-1a was evaluated by Westernblot. After intracavernous injection of pSV-Luc or pRTP801-Luc, gene expression was evaluated by luciferase assay,and the gene expression area was evaluated by immunohistochemistry.Results. HIF-1a was up-regulated in the corpus cavernosum of hypercholesterolemic rats. Although pSV-Luc didnot induce gene expression in either the control or the cholesterol group, pRTP801-Luc significantly induced geneexpression in the cholesterol group and resulted in higher luciferase activity than did pSV-Luc up to 14 days afterinjection. Immunohistochemistry showed that the gene expression area was also greater in the pRTP801-Luc groupthan in the pSV-Luc group, but the difference was not as great as that in luciferase activity. This suggests thatpRTP801-Luc exerts its effect mainly by inducing promoter activity under hypoxia, not by increasing the number oftransfected cells.Conclusion. The RTP801 promoter-driven gene expression system increased gene expression in the corpus caver-nosum tissue of rats with cholesterol-induced ED. This may be a useful system for the development of gene therapyin vasculogenic ED. Lee M, Ryu J-K, Piao S, Choi MJ, Kim HA, Zhang L-W, Shin H-Y, Jung HI, Kim I-H,Kim SW, and Suh J-K. Efficient gene expression system using the RTP801 promoter in the corpuscavernosum of high-cholesterol diet-induced erectile dysfunction rats for gene therapy. J Sex Med 2008;5:1355–1364.

Key Words. Gene Therapy; Hypoxia-Inducible Factor-1a; Hypercholesterolemia; Erectile Dysfunction; CorpusCavernosum; Vector

Introduction

D espite the introduction of oral phosphodi-esterase (PDE)-5 inhibitors in the treatment

of erectile dysfunction (ED), new therapeutic

strategies are warranted. Gene therapy representsan exciting new option for the treatment of ED,because it may provide definite advantages forthose who respond poorly to oral PDE-5 inhibi-tors because of severe cardiovascular disease, those

1355

© 2008 International Society for Sexual Medicine J Sex Med 2008;5:1355–1364

with diabetes or radical prostatectomy, those whotake nitrates, those who do not allow for planningbefore intercourse, and those who need a treat-ment modality that can cure the underlying disor-der in the corpus cavernosum [1–6].

Gene therapy for ED has been developed withvarious therapeutic genes in several pathologicconditions. However, most of these studies, exceptfor a clinical trial with the naked DNA plasmidcarrying the human cDNA encoding maxi-Kchannel [7], remain at the preclinical level, eventhough the results are impressive and encouraging.One of the major reasons for this is that the viralvectors used in the studies have potential risks,such as immunogenicity, that limit their applica-tion in a clinical setting [8,9]. In contrast, nakedplasmid DNA is safe, but it has a low transfectionefficiency in most cell types.

The application of gene therapy in ED requiresa higher level of safety and more knowledge ofsecure and efficacious systems for gene transferthan the use of gene therapy for life-threateningdisease. A gene expression system is needed thatwill induce the expression of the therapeutic genesmainly in the diseased target tissues and minimizethe expression in normal tissues to avoid possibleside effects. For example, in gene therapy withgrowth factor genes such as vascular endothelialgrowth factor (VEGF), unregulated VEGFexpression can induce endothelial cell-derivedintramural vascular tumors or promote the ruptureof atherosclerotic plaque [10,11]. Also, VEGFexpression has been shown to have physiologiceffects in normal, nonischemic tissues [11]. Thus,gene expression must be tightly regulated. The useof tissue-specific promoters or enhancers in theexpression cassettes has recently been introducedas a promising approach for targeting gene expres-sion in diseased tissues [12].

Cavernous hypoxia caused by vascular riskfactors or cavernous nerve injury has been impli-cated as one of the most important contributingfactors in the pathophysiology of ED [13,14]. Ofthose vascular risk factors, hypercholesterolemia iswell known to have a powerful effect on the devel-opment of both angiopathy and ED [15–19]. Pre-viously, Azadzoi et al. [20,21] reported a decreasein intracavernous blood flow and intracavernousoxygen tension in an animal model of vasculogenicED induced by balloon de-endothelialization ofiliac arteries and a high-cholesterol diet. Recently,the RTP801 promoter has been used as an effec-tive regulatory system for VEGF gene therapy[22,23], and the induction of these regulatory

elements is mediated by hypoxia-induciblefactor-1 (HIF-1) [24]. Here we report on the effec-tiveness of the RTP801 promoter in a targetinggene expression system for gene therapy in ananimal model of vasculogenic ED induced byhigh-cholesterol diet.

Methods

Animals and Study DesignA total of 80 male, 2-month-old Sprague–Dawleyrats were used in this study. The experiments per-formed were approved by the Institutional AnimalCare and Use Subcommittee of our university.The 18 control animals were fed a normal diet,and the remaining 62 experimental animals werefed a diet containing 4% cholesterol and 1% cholicacid (Hansam Biotech Co., Gyeonggi, Korea) for3 months. In six controls and six hypercholester-olemic rats, erectile function was evaluated bycavernous nerve electrical stimulation. Cavernoustissue specimens from the remaining 12 controlsand 56 hypercholesterolemic rats were used forbiochemical study, including luciferase assay andWestern blot assay, and histologic evaluation.Blood was extracted from the carotid artery or bydirect cardiac puncture, and total cholesterol con-centrations were measured.

Measurement of Erectile FunctionFor electrically stimulated penile erections, abipolar platinum electrode was placed around thecavernous nerve. Stimulation parameters were 5 Vat a frequency of 12 Hz, with square-wave dura-tion of 1 ms for 1 minute. The ratio of maximalintracavernous pressure (ICP) to mean arterialpressure (MAP) at the peak erectile response wascalculated to control for variations in systemicblood pressure. Total ICP was determined by thearea under the curve from the beginning of thecavernous nerve stimulation until ICP returned tobaseline, and tumescence slope for ICP to reach80% of maximal ICP (S80) was calculated.

Western BlotThe expression of hypoxia-inducible factor(HIF)-1a protein in control and hypercholester-olemic rat corpus cavernosum was detected byWestern blot (N = 4, respectively). Equal amountsof protein (60 mg/lane) were electrophoresed on8% sodium dodecyl sulfate (SDS)-polyacrylamidegels, transferred to nitrocellulose membranes, andprobed with antibody to HIF-1a (1:200; Santa

1356 Lee et al.

J Sex Med 2008;5:1355–1364

Cruz Biotechnology, Santa Cruz, CA, USA) orb-actin (Abcam, Cambridge, UK). The loadedamounts of the extracts in SDS-polyacrylamidegels were in the range of linearity between thesignal intensity and the loaded amount. Theresults were quantified by densitometry.

Preparation of PlasmidsThe construction of pRTP801-Luc was describedpreviously [22]. pSV-Luc (pGL3-promoter) waspurchased from Promega (Madison, WI, USA).pSV-Luc and pRTP801-Luc were introduced intoEscherichia coli strain JM109 (Promega) and werepurified by use of Qiagen Plasmid Maxi Kits(Qiagen, Valencia, CA, USA). The purity of theplasmids was certified by measuring the opticaldensity (OD)260/OD280 ratio. The concentration ofplasmid DNA was determined by using 1(OD260) = 50 mg of DNA. Plasmids were stored at(-20°C until used. In pRTP801-Luc, the expres-sion of luciferase is under the control of theRTP801 promoter, while the SV40 promoter con-trols the expression in pSV40-Luc. RTP801 wasfirst identified as a hypoxia-responsive gene [24].The RTP801 promoter has an HIF-1a bindingsite in its proximal region. It was previously provedthat the RTP801 promoter was sharply inducedunder hypoxia, which was mediated by the bindingof HIF-1 [24].

In Vitro TransfectionTissue samples of corpus cavernosum wereobtained from a 33-year-old potent patient, whounderwent corporoplasty for congenital penilecurvature. This human study was approved bythe institutional review board of our university.Primary cultures of human penile corpus caverno-sum smooth muscle cells were established, as pre-viously described [25,26], and were used at earlypassage (P2–P4) for all experiments. Human corpuscavernosum smooth muscle cells were maintainedin Dulbecco modified eagle medium (DMEM)supplemented with 10% fetal bovine serum in a 5%CO2 incubator. For the transfection studies, thecells were seeded at a density of 2.5 ¥ 105 cells/wellin 6-well plates 24 hours before transfection. After24 hours, the cells were washed twice with serum-free DMEM medium, and 2 mL of fresh serum-free medium was added. Polyethylenimine (PEI,25 kDa) was used as a gene carrier. PEI-plasmidcomplexes were prepared at a 5:1 ratio of N (nitro-gen of PEI) to P (phosphate of DNA) on the basisof previous reports [27–31]. The PEI-plasmidcomplex was added to each well. The amount of

plasmid was fixed at 2 mg/well. The cells were thenincubated for 4 hours at 37°C in a 5% CO2 incu-bator. After 4 hours, the transfection mixtures wereremoved, and 2 mL of fresh DMEM medium con-taining fetal bovine serum was added. The cellswere incubated at the desired concentration ofoxygen (normoxia, 20% O2, 5% CO2, 75% N2;hypoxia, 1% O2, 5% CO2, 94% N2) for 20 hours.

In Vivo Gene Delivery to the Corpus CavernosumThe rats from the cholesterol group and their age-matched controls were anesthetized with ketamine(100 mg/kg) and xylazine (5 mg/kg) intramuscu-larly and were placed on a thermoregulated surgi-cal table. With the use of sterile technique, thepenile skin was incised and the tunica albugineawas exposed. The pSV-Luc (100 mg/0.1 mL)or pRTP801-Luc (100 mg/0.1 mL) vector wasinjected into the midportion of the corpus caver-nosum of the control or the hypercholesterolemicrats, and the needle was left in place for 2 minutesto allow the gene to diffuse throughout the cav-ernous space. According to the results of our pre-vious study, blood drainage via the dorsal veins washalted by circumferential compression of the penisat the base with an elastic band immediately beforeinjection, and the compression was released at 30minutes after the injection [26]. The incision wasclosed with 6-0 Vicryl (polyglactin 910) sutures. At7 days after injection, the penis was harvested andgene expression was measured by luciferase assay(N = 4 per group). Based on this initial result,a separate group of hypercholesterolemic ratsreceived pSV-Luc (100 mg/0.1 mL) or pRTP801-Luc (100 mg/0.1 mL). At 3, 7, 14, or 28 days afterinjection, the penis was harvested for luciferaseassay (N = 4 per group and per each time point).

Luciferase AssayAfter transfection, the cells were washed withphosphate-buffered saline, and 150 mL of reporterlysis buffer (Promega) was added to each well.After 15 minutes of incubation at room tempera-ture, the cells were harvested and transferred tomicrocentrifuge tubes. After 15 seconds of vortex-ing, the cells were centrifuged at 11,000 rpm for3 minutes. For in vivo study, the tissues werehomogenized in reporter lysis buffer. The homo-genates were centrifuged at 11,000 rpm for 3minutes. The extracts were transferred to freshtubes. The protein concentrations of the extractswere determined by using a bicinchoninic acid(BCA) protein assay kit (Pierce, Iselin, NJ, USA).Luciferase activity was measured in terms of rela-

Effectiveness of the RTP801 promoter in Gene Therapy 1357

J Sex Med 2008;5:1355–1364

tive light units (RLUs) by using a 96-wellplate luminometer (Berthold Detection SystemGmbH, Pforzheim, Germany). The final valuesof luciferase were reported in terms of RLU permilligram of total protein.

Immunofluorescence Double Staining inthe Corpus CavernosumRats from the cholesterol group were injectedwith pSV-Luc or pRTP801-Luc (100 mg/0.1 mL,N = 6, respectively) intracavernously, and corpuscavernosum tissue was harvested 3 days after injec-tion. The frozen tissue sections (10 mm) were incu-bated with antibody to luciferase (1:50; ZymedLaboratories, San Francisco, CA, USA), factorVIII (1:50; DakoCytomation, Glostrup, Den-mark), or smooth muscle a-actin (Abcam, 1:25) at4°C overnight. Factor VIII was used as a markerfor endothelial cells, while smooth muscle a-actinwas used to label smooth muscle cells. Controlsections were incubated without the primary anti-body at this step. After several washes withphosphate-buffered saline, the sections were incu-bated with fluorescein isothiocyanate (FITC)- ortetramethyl rhodarnine isothiocyanate (TRITC)-conjugated secondary antibodies for 2 hours atroom temperature. Signals were visualized, anddigital images were obtained with an Apotomemicroscope (Zeiss, Göttingen, Germany). Mul-tiple sections were carefully examined, digitized,and merged after image capture to identify thegreatest area of congruence among luciferase-positive cells and endothelial or smooth musclecells. For estimation of transfection efficacy,luciferase-positive areas were analyzed with animage analyzer system (National Institutes ofHealth [NIH] Image J 1.34, http://rsb.info.nih.gov/ij/index.html). The relative area in thecorpus cavernosum, endothelial, or smooth musclecells was determined and expressed as a percentageof the total each area as follows: (luciferase-positive area/total area) ¥ 100.

Statistical AnalysisResults were expressed as means � standard devia-tions. Mann–Whitney U-tests were used to evalu-ate whether differences between groups weresignificant. Probability values less than 5% wereconsidered significant.

Results

High-Cholesterol Diet-Induced ED ModelThe serum total cholesterol concentrations ofanimals fed with the high-cholesterol diet

(234.2 � 75.3 mg/dL) were significantly higherthan the concentrations in the controls (94.3 �20.2 mg/dL, P < 0.01). All erectile function vari-ables assessed, such as the ratio of maximal ICP toMAP (0.89 � 0.14 vs. 0.46 � 0.12), total ICP(47.3 � 9.4 vs. 24.4 � 5.2), and slope (0.022 �0.006 vs. 0.011 � 0.004 cm H2O/sec), were signifi-cantly lower in the hypercholesterolemic rats thanin the controls (P < 0.01, respectively). No detect-able differences were found in resting ICP or MAPbetween the groups (data not shown).

Induction of HIF-1a in the Corpus CavernosumWestern blot assay with antibody to HIF-1ashowed that HIF-1a was induced in the corpuscavernosum of the hypercholesterolemic rats(Figure 1A). By contrast, no induction in expres-sion of the internal control, b-actin, was shown(Figure 1A). The densitometry results showedthat the cholesterol group expressed more than100% more HIF-1a than did the control group(Figure 1B). This experiment suggests that thehypoxia-inducible gene expression system maybe applicable for gene therapy in the corpuscavernosum.

In Vitro Evaluation of the RTP801 PromoterTo evaluate luciferase expression under hypoxia ornormoxia in vitro, pSV-Luc and pRTP801-Lucwere transfected into human corpus cavernosumsmooth muscle cells. The structures of the plas-mids are presented in Figure 2. The transfectionswere performed with PEI as a gene carrier. In thecells transfected with pSV-Luc, luciferase expres-sion was not induced by incubation under hypoxia(Figure 3). By contrast, pRTP801-Luc inducedexpression in the transfected cells under hypoxia(Figure 3).

In Vivo Evaluation of the RTP801 PromoterSeven days after the intracavernous injection ofpRTP801-Luc or pSV-Luc, tissues were harvestedand gene expression was evaluated by luciferaseassay. pSV-Luc did not show any induction inexpression in the cholesterol group compared tothe control group. However, pRTP801-Lucresulted in higher gene expression in the choles-terol group (Figure 4A). The gene expression levelin the cholesterol group was sixfold that of thecontrol group. The duration of gene expressionwas also evaluated. pRTP801-Luc showed highergene expression than pSV-Luc for more than14 days (Figure 4B). By 28 days, however, geneexpression with the two vectors was similar.

1358 Lee et al.

J Sex Med 2008;5:1355–1364

Evaluation of the Transfected AreaTypical immunostaining results are presented inFigure 5A. At 3 days after intracavernous genetransfer of pRTP801-Luc in hypercholesterolemic

rats, the gene expression area in the corpus caver-nosum was slightly larger than that in the groupinjected with pSV-Luc. The expression areaof the pRTP801-Luc group was approximately50% greater than that of the pSV-Luc group(Figure 5B). We also performed immunofluores-cence double staining to identify which subpopu-lation of corpus cavernosum cells was transfected.After intracavernous injection of pSV-Luc orpRTP801-Luc, the luciferase expression over-lapped cavernous endothelial cells and smoothmuscle cells (Figure 5A). However, the geneexpression area in the endothelial cells or smoothmuscle cells did not differ significantly betweenpSV-Luc and pRTP801-Luc groups (Figure 5B).This result was in contrast with the results of theluciferase assay. Three days after the intracavern-ous administration of each gene, the luciferaseassay showed that the expression induced by

Figure 1 Hypoxia-inducible factor (HIF)-1a expression inthe corpus cavernosum of control and hypercholesterolemicrats. (A) The expression of HIF-1a protein in control andhypercholesterolemic (cholesterol) rat corpus cavernosumwas detected by Western blot assay (N = 4, respectively).Equal amounts of protein (60 mg/lane) were electrophore-sed on 8% sodium dodecylsulfate-polyacrylamide gels,transferred to nitrocellulose membranes, and probed withantibody to HIF-1a or b-actin. (B) The results were quanti-fied by densitometry. The data were expressed as the meanvalues (�standard deviation) of four experiments. *P < 0.05vs. control group.

Figure 2 The structures of the pSV-Luc and pRTP801-Luc transfectionplasmids.

Figure 3 Luciferase expression driven by pSV-Luc orpRTP801-Luc in human corpus cavernosum smoothmuscle cells. Plasmid-polyethylenimine complexes wereprepared with pSV-Luc or pRTP801-Luc. The complexeswere transfected into human corpus cavernosum smoothmuscle cells. The cells were exposed to normoxia orhypoxia for 20 hours. Luciferase expression was measuredby luciferase assay. The data were expressed as the meanvalues (�standard deviation) of four experiments. *P < 0.01vs. normoxia group. RLU = relative light unit.

Effectiveness of the RTP801 promoter in Gene Therapy 1359

J Sex Med 2008;5:1355–1364

pRTP801-Luc was more than eight times as muchas that induced by pSV-Luc (Figure 4B).

Discussion

In the present study, the ratio of maximal ICP toMAP was significantly lower in the hypercholes-terolemic rats than in the controls. However, themaximal ICP/MAP ratio in the control rats (0.89)was somewhat higher than the ratio by otherinvestigators [32,33]. The erectile function vari-ables during electrical stimulation of the cavernousnerve can be affected by stimulation parameters orby the stimulation device used. So, we think thatthis disparity may result from differences in stimu-lation parameters or the stimulation device used.HIF-1a expression was induced in the corpus cav-ernosum tissue of rats with ED induced by a high-cholesterol diet. This implies that cavernoustissue from hypercholesterolemic rats may havelow oxygen tension, a condition under which thehypoxia-inducible gene expression system mayactively operate. As a result, the RTP801 promoterexpression system proved to dramatically increasereporter gene expression in the corpus cavernosumof hypercholesterolemic ED rats.

Previously, we developed hypoxia-induciblesystems by using various DNA elements suchas the RTP801 promoter and the erythropoietin(Epo) 3’-untranslated region (UTR) [22,23,34].The RTP801 promoter has strong promoter activ-ity in cells such as endothelial cells, smooth musclecells, kidney cells, and hepatocytes under hypoxia[22]. The in vitro transfection assay in the presentstudy showed that the RTP801 promoter hadstronger promoter activity in human corpus caver-nosum smooth muscle cells than did the SV40promoter. Furthermore, gene expression by theRTP801 promoter was significantly induced underhypoxia in vitro (Figure 3). The direct injection ofpRTP801-Luc into the corpus cavernosum alsoresulted in greater gene expression in the choles-terol group than in the control group. The geneexpression after intracavernous delivery ofpRTP801-Luc into the hypercholesterolemic ratspeaked at the earliest time point at which it wasassessed (3 days), and the higher expression wasmaintained for at least 2 weeks (Figure 4). Histo-logically, the luciferase-positive area 3 days afterintracavernous administration of pRTP801-Lucwas 50% greater than that of pSV-Luc (Figure 5).However, luciferase activity measured at this timepoint in hypercholesterolemic rats injected withpRTP-Luc wasapproximately eight times as much

Figure 4 Luciferase expression by pSV-Luc or pRTP801-Luc in the corpus cavernosum of control and hypercholes-terolemic rats. (A) pSV-Luc or pRTP801-Luc was injectedinto the midportion of the corpus cavernosum of control orhypercholesterolemic (cholesterol) rats. The penis was har-vested and gene expression was measured by luciferaseassay 7 days after injection. The data were expressed asmean values (�standard deviation) for N = 4 animalsper group. *P < 0.01 vs. control group. (B) pSV-Luc orpRTP801-Luc was injected into the corpus cavernosum ofhypercholesterolemic rats. The penis was harvested andgene expression was measured by luciferase assay 3, 7,14, and 28 days after injection. The data were expressed asmean values (�standard deviation) for N = 4 animals pergroup and per each time point. *P < 0.01 vs. the pSV-Lucgroup. RLU = relative light unit.

1360 Lee et al.

J Sex Med 2008;5:1355–1364

as that in rats injected with pSV-Luc. This discrep-ancy suggests that the RTP801 promoter may exertits effect predominantly by inducing promoteractivity under hypoxia, and not by increasing trans-fection efficiency, i.e., the number of transfectedcells, which in turn resulted in a dramatic increase

in protein expression in the transfected cells.Therefore, this promoter system may not be appli-cable to genes of nonsecreting therapeutic proteins,because it would not increase extracellular concen-trations of the proteins. However, in genes ofsecreting therapeutic proteins, such as VEGF,

Figure 5 Luciferase expression area in the corpus cavernosum, endothelial, or smooth muscle cells. (A) Hypercholester-olemic rats were injected with pSV-Luc or pRTP801-Luc intracavernously, and the corpus cavernosum tissue was harvested3 days after injection. The frozen tissue sections were stained with antibody to luciferase (green), factor VIII (red), or smoothmuscle a-actin (red). Immunofluorescence for luciferase was found in both endothelial cells and in the smooth muscle cells(yellow). Original magnification 100¥. (B) Luciferase-positive areas were analyzed with an image analyzer system (NIHImage J 1.34). The data were expressed as mean values (�standard deviation) for N = 6 animals per group. *P < 0.01 vs.pSV-Luc group. 2nd Ab = secondary antibody control; Luc = luciferase.

Effectiveness of the RTP801 promoter in Gene Therapy 1361

J Sex Med 2008;5:1355–1364

angiopoietin-1, and brain-derived neurotrophicfactor, the RTP801 promoter can increase theextracellular therapeutic protein level, especially inischemic tissues such as the corpus cavernosum.

The safety and efficacy of this hypoxia-inducible RTP801 promoter system may beimproved for application in the future treatment ofhuman ED. For tighter regulation of the desiredgene expression, a hypoxia-responsive RNA ele-ment such as Epo 3′-UTR can be applied as apost-transcriptional regulation system [34]. Also,an oxygen-dependent degradation domain will beuseful for rapid degradation of therapeutic proteinunder normoxia or in normal tissue [35]. In ourprevious study, we have evaluated the efficiency ofa nonviral gene delivery carrier, water-solublelipopolymer (WSLP). It has low toxicity and hightransfection efficiency both in rat corpus caverno-sum tissue and human corpus cavernosum smoothmuscle cells [26]. Therefore, it may be a usefulstrategy to combine hypoxia-inducible RTP801promoter and WSLP, a gene delivery carrier, forthe development of a more efficient gene expres-sion system to treat vasculogenic ED.

Conclusion

The RTP801 promoter-driven gene expressionsystem increased gene expression in the corpuscavernosum tissue of rats with ED induced by ahigh-cholesterol diet. The gene expression systemworked predominantly by enhancing promoteractivity under hypoxia, not by increasing thenumber of transfected cells. This gene expressionsystem may be useful for the development of genetherapy in vasculogenic ED.

Acknowledgments

This article was supported by grant no. R01-2005-000-10411-0 from the Basic Research Program of the KoreaScience and Engineering Foundation (Jun-Kyu Suh), bygrant no. 410053 from Korea Research Foundation(Lu-Wei Zhang and Jun-Kyu Suh), and by the BrainKorea 21 Project in 2006 (Shuguang Piao, Ji-Kan Ryu,and Jun-Kyu Suh). The authors thank Jennifer Scalesfor help in preparing the manuscript.

Corresponding Author: Jun-Kyu Suh, MD, PhD,Department of Urology, Inha University Hospital,7-206, 3rd St. Shinheung-Dong Jung-Gu, Incheon,Korea 400-711. Tel: 82-32-890-3441; Fax: 82-32-890-3097; E-mail: [email protected]

Conflict of Interest: None declared.

Statement of Authorship

Category 1(a) Conception and Design

Minhyung Lee; In-Hoo Kim; Sung Wan Kim; Jun-Kyu Suh

(b) Acquisition of DataMinhyung Lee; Ji-Kan Ryu; Shuguang Piao; Min JiChoi; Hyun Ah Kim; Lu-Wei Zhang; Hwa-YeanShin; Haeng In Jung

(c) Analysis and Interpretation of DataMinhyung Lee; Ji-Kan Ryu; In-Hoo Kim; SungWan Kim; Jun-Kyu Suh

Category 2(a) Drafting the Article

Minhyung Lee(b) Revising It for Intellectual Content

Minhyung Lee; Ji-Kan Ryu; Jun-Kyu Suh

Category 3(a) Final Approval of the Completed Article

Jun-Kyu Suh

References

1 Goldstein I, Lue TF, Padma-Nathan H, Rosen RC,Steers WD, Wicker PA. Oral Sildenafil in the treat-ment of erectile dysfunction. Sildenafil StudyGroup. N Engl J Med 1998;338:1397–404.

2 Rendell MS, Rajfer J, Wicker PA, Smith MD,Sildenafil Diabetes Study Group. Sildenafil fortreatment of erectile dysfunction in men withdiabetes: A randomized controlled trial. JAMA1999;281:421–6.

3 Setter SM, Iltz JL, Fincham JE, Campbell RK,Baker DE. Phosphodiesterase 5 inhibitors for erec-tile dysfunction. Ann Pharmacother 2005;39:1286–95.

4 Martinez-Jabaloyas JM, Gil-Salom M, Villamon-Fort R, Pastor-Hernandez F, Martinez-Garcia R,Garcia-Sisamon F. Prognostic factors for responseto sildenafil in patients with erectile dysfunction.Eur Urol 2001;40:641–6.

5 Gresser U, Gleiter CH. Erectile dysfunction:Comparison of efficacy and side effects of thePDE-5 inhibitors sildenafil, vardenafil andtadalafil—Review of the literature. Eur J Med Res2002;7:435–46.

6 Herrmann HC, Levine LA, Macaluso J Jr, Walsh M,Bradbury D, Schwartz S, Mohler ER III, KimmelSE. Can atorvastatin improve the response tosildenafil in men with erectile dysfunction not ini-tially responsive to sildenafil? Hypothesis and pilottrial results. J Sex Med 2006;3:303–8.

7 Melman A, Bar-Chama N, McCullough A, DaviesK, Christ G. hMaxi-K gene transfer in males with

1362 Lee et al.

J Sex Med 2008;5:1355–1364

erectile dysfunction: results of the first human trial.Hum Gene Ther 2006;18:1165–76.

8 Temin HM. Safety considerations in somatic genetherapy of human disease with retrovirus vectors.Hum Gene Ther 1990;1:111–23.

9 Kabanov AV, Kabanov VA. DNA complexes withpolycations for the delivery of genetic material intocells. Bioconjug Chem 1995;6:7–20.

10 Springer ML, Chen AS, Kraft PE, Bednarski M,Blau HM. VEGF gene delivery to muscle: Potentialrole for vasculogenesis in adults. Mol Cell1998;2:549–58.

11 Lee RJ, Springer ML, Blanco-Bose WE, Shaw R,Ursell PC, Blau HM. VEGF gene delivery to myo-cardium: Deleterious effects of unregulated expres-sion. Circulation 2000;102:898–901.

12 Somia N, Verma IM. Gene therapy: Trials andtribulations. Nat Rev Genet 2000;1:91–9.

13 Moreland RB. Is there a role of hypoxemia in penilefibrosis: A viewpoint presented to the Society for theStudy of Impotence. Int J Impot Res 1998;10:113–20.

14 Vignozzi L, Filippi S, Morelli A, Ambrosini S,Luconi M, Vannelli GB, Donati S, Crescioli C,Zhang XH, Mirone V, Forti G, Maggi M. Effectofchronic tadalafil administration on penile hypoxiainduced by cavernousneurotomy in the rat. J SexMed 2006;3:419–31.

15 Ryu JK, Cho CH, Shin HY, Song SU, Oh SM, LeeM, Piao S, Han JY, Kim IH, Koh GY, Suh JK.Combined angiopoietin-1 and vascular endothelialgrowth factor gene transfer restores cavernousangiogenesis and erectile function in a rat model ofhypercholesterolemia. Mol Ther 2006;13:705–15.

16 Ryu JK, Shin HY, Song SU, Oh SM, Piao S, HanJY, Park KW, Suh JK. Downregulation of angio-genic factors and their downstream target moleculesaffects the deterioration of erectile function in a ratmodel of hypercholesterolemia. Urology 2006;67:1329–34.

17 Xie D, Odronic SI, Wu F, Pippen AM, DonatucciCF, Annex BH. A mouse model ofhypercholesterolemia-induced erectile dysfunction.J Sex Med 2007;4:898–907.

18 Xie D, Pippen AM, Odronic SI, Annex BH,Donatucci CF. Intracavernosal basic fibroblastgrowth factor improves vasoreactivity in the hyper-cholesterolemic rabbit. J Sex Med 2006;3:223–32.

19 Park K, Kim SW, Rhu KS, Paick JS. Chronicadministration of an oral Rho kinase inhibitorprevents the development of vasculogenic erectiledysfunction in a rat model. J Sex Med 2006;3:996–1003.

20 Azadzoi KM, Park K, Andry C, Goldstein I, SirokyMB. Relationship between cavernosal ischemia andcorporal veno-occlusive dysfunction in an animalmodel. J Urol 1997;157:1011–7.

21 Park K, Siroky MB, Goldstein I. Relationshipbetween atherosclerosis-induced cavernosal hypoxia

and erectile dysfunction in the rabbit. Int J ImpotRes 1996;8:104 (Abstract A22).

22 Lee M, Bikram M, Oh S, Bull DA, Kim SW. Sp1-dependent regulation of the RTP801 promoter andits application to hypoxia-inducible VEGF plasmidfor ischemic disease. Pharm Res 2004;21:736–41.

23 Lee M, Lee ES, Kim YS, Choi BH, Park SR, ParkHS, Park HC, Kim SW, Ha Y. Ischemic injury-specific gene expression in the rat spinal cord injurymodel using hypoxia-inducible system. Spine2005;30:2729–34.

24 Shoshani T, Faerman A, Mett I, Zelin E, Tenne T,Gorodin S, Moshel Y, Elbaz S, Budanov A, ChajutA, Kalinski H, Kamer I, Rozen A, Mor O, Keshet E,Leshkowitz D, Einat P, Skaliter R, Feinstein E.Identification of a novel hypoxia-inducible factor1-responsive gene, RTP801, involved in apoptosis.Mol Cell Biol 2002;22:2283–93.

25 Moreland RB, Traish A, McMillin MA, Smith B,Goldstein I, Saenz de Tejada I. PGE1 suppresses theinduction of collagen synthesis by transforminggrowth factor-beta1 in human corpus cavernosumsmooth muscle. J Urol 1995;153:826–34.

26 Lee M, Ryu JK, Oh SM, Lee E, Shin HY, Song SU,Kim SW, Suh JK. Water-soluble lipopolymer as agene carrier to corpus cavernosum. Int J Impot Res2005;17:326–34.

27 Abdallah B, Hassan A, Benoist C, Goula D, Behr JP,Demeneix BA. A powerful nonviral vector for invivo gene transfer into the adult mammalian brain:Polyethylenimine. Hum Gene Ther 1996;7:1947–54.

28 Lemkine GF, Goula D, Becker N, Paleari L,Levi G, Demeneix BA. Optimisation ofpolyethylenimine-based gene delivery to mousebrain. J Drug Target 1999;7:305–12.

29 Turunen MP, Hiltunen MO, Ruponen M,Virkamaki L, Szoka FC Jr, Urtti A, Yla-Herttuala S.Efficient adventitial gene delivery to rabbit carotidartery with cationic polymer-plasmid complexes.Gene Ther 1999;6:6–11.

30 Nguyen HK, Lemieux P, Vinogradov SV, GebhartCL, Guerin N, Paradis G, Bronich TK, AlakhovVY, Kabanov AV. Evaluation of polyether-polyethyleneimine graft copolymers as gene transferagents. Gene Ther 2000;7:126–38.

31 Lee M, Rentz J, Han SO, Bull DA, Kim SW. Water-soluble lipopolymer as an efficient carrier for genedelivery to myocardium. Gene Ther 2003;10:585–93.

32 Giuliano F, Rampin O, Bernabe J, Rousseau JP.Neural control of penile erection in the rat. J AutonNerv Syst 1995;55:36–44.

33 Bivalacqua TJ, Usta MF, Kendirci M, Pradhan L,Alvarez X, Champion HC, Kadowitz PJ, HellstromWJ. Superoxide anion production in the rat penisimpairs erectile function in diabetes: Influence ofin vivo extracellular superoxide dismutase genetherapy. J Sex Med 2005;2:187–97.

Effectiveness of the RTP801 promoter in Gene Therapy 1363

J Sex Med 2008;5:1355–1364

34 Lee M, Choi D, Choi MJ, Jeong JH, Kim WJ, Oh S,Kim YH, Bull DA, Kim SW. Hypoxia-induciblegene expression system using the erythropoietinenhancer and 3’-untranslated region for the VEGFgene therapy. J Control Release 2006;115:113–9.

35 Lee JW, Bae SH, Jeong JW, Kim SH, Kim KW.Hypoxia-inducible factor (HIF-1)alpha: Its proteinstability and biological functions. Exp Mol Med2004;36:1–12.

1364 Lee et al.

J Sex Med 2008;5:1355–1364